Executive Summary
CytoAtlas is a comprehensive computational resource that maps cytokine and secreted protein signaling activity across ~29 million human cells and ~31,000 bulk RNA-seq samples from six independent datasets: two bulk RNA-seq resources (GTEx, TCGA) and four single-cell compendia (CIMA, Inflammation Atlas, scAtlas, parse_10M) spanning healthy donors, inflammatory diseases, cancers, and cytokine perturbations. The system uses linear ridge regression against experimentally derived signature matrices to infer activity — producing fully interpretable, conditional z-scores rather than black-box predictions.
Key results:
- 1,213 signatures (43 CytoSig + 1,170 SecAct), plus 178 cell-type-specific LinCytoSig variants, validated across 6 independent datasets
- Spearman correlations reach ρ=0.6–0.9 for well-characterized cytokines (IL1B, TNFA, VEGFA, TGFB family)
- Cross-dataset consistency demonstrates signatures generalize across CIMA, Inflammation Atlas Main, scAtlas, GTEx, and TCGA
- SecAct achieves the highest median correlations in 5 of 6 datasets (independence-corrected median ρ=0.31–0.46)
Table of Contents
1. System Architecture and Design Rationale
1.1 Architecture and Processing [Detailed Architecture →]
Linear interpretability over complex models. Ridge regression (L2-regularized linear regression) was chosen deliberately over methods like autoencoders, graph neural networks, or foundation models. The resulting activity z-scores are conditional on the specific genes in the signature matrix, meaning every prediction can be traced to a weighted combination of known gene responses. This is critical for biological interpretation — a scientist can ask “which genes drive the IFNG activity score in this sample?” and get a direct answer.
Reproducibility through separation of concerns. The system is divided into independent components, each chosen for the constraints of HPC/SLURM infrastructure:
| Component | Technology | Purpose | Rationale |
|---|---|---|---|
| Pipeline | Python + CuPy (GPU) | Activity inference | 10–34x speedup over NumPy; batch-streams H5AD files (500K–1M cells/batch) with projection matrix held on GPU; automatic CPU fallback when GPU unavailable |
| Storage | DuckDB (3 databases, 68 tables) | Columnar analytics | Single-file databases require no server — essential on HPC where database servers are unavailable; each database regenerates independently without affecting others |
| API | FastAPI (262 endpoints) | RESTful data access | Async I/O for concurrent DuckDB queries; automatic OpenAPI documentation; Pydantic request validation; lifespan management for resource initialization |
| Frontend | React 19 + TypeScript | Interactive exploration (12 pages) | Migrated from 25K-line vanilla JS SPA to 11.4K lines (54% reduction) with type safety, component reuse, and lazy-loaded routing |
Processing scale. Ridge regression (λ=5×105) is applied using secactpy.ridge() against each signature matrix. For single-cell data, expression is first aggregated to pseudobulk (donor or donor×celltype level), then genes are intersected with the signature matrix (CytoSig: ~4,860 genes; SecAct: ~7,450 genes). The resulting z-scored activity coefficients are compared to target gene expression via Spearman correlation across donors.
| Dataset | Cells/Samples | Processing Time | Hardware |
|---|---|---|---|
| GTEx | 19,788 bulk samples | ~10min | A100 80GB |
| TCGA | 11,069 bulk samples | ~10min | A100 80GB |
| CIMA | 6.5M cells | ~2h | A100 80GB |
| Inflammation Atlas (main/val/ext) | 6.3M cells | ~2h | A100 80GB |
| scAtlas Normal | 2.3M cells | ~1h | A100 80GB |
| scAtlas Cancer | 4.1M cells | ~1h | A100 80GB |
| parse_10M | 9.7M cells | ~3h | A100 80GB |
Total: ~29M single cells + ~31K bulk RNA-seq samples, processed through ridge regression against 3 signature matrices (CytoSig: 43 cytokines, LinCytoSig: 178 cell-type-specific, SecAct: 1,170 secreted proteins). Processing Time = wall-clock time for full activity inference on a single NVIDIA A100 GPU (80 GB VRAM). For bulk datasets (GTEx/TCGA), ridge regression is applied with within-tissue/within-cancer mean centering to remove tissue-level variation. See Section 2.1 for per-dataset details.
1.2 Validation Strategy
CytoAtlas validates at four aggregation levels, each testing whether predicted activity correlates with target gene expression (Spearman ρ) across independent samples:
| Level | Description | Datasets | Report Section |
|---|---|---|---|
| Donor pseudobulk | One value per donor, averaging across cell types | CIMA, Inflammation Atlas Main, scAtlas Normal/Cancer | §4.1, §4.3 |
| Donor × cell-type | Stratified by cell type within each donor | CIMA, Inflammation Atlas Main, scAtlas Normal/Cancer | §4.7 |
| Per-tissue / per-cancer | Median-of-medians across tissues or cancer types | GTEx (29 tissues), TCGA (33 cancer types) | §4.2, §4.3 |
| Cross-platform | Bulk vs pseudobulk concordance per tissue/cancer | GTEx vs scAtlas Normal, TCGA vs scAtlas Cancer | §4.4 |
All statistics use independence-corrected values — preventing inflation from repeated measures across tissues, cancer types, or cell types. CytoSig vs SecAct comparisons use Mann-Whitney U (total) and Wilcoxon signed-rank (32 matched targets) with BH-FDR correction. See Section 3.3 for the validation philosophy and Section 4 for full results.
Why independence correction matters: Pooling across tissues or cancer types inflates correlations through confounding. For example, GTEx pooled CytoSig median ρ (0.211) is 40% higher than the independence-corrected by-tissue value (0.151); SecAct shows +30% inflation (0.394 vs 0.304). All results in this report use the corrected values. For a detailed comparison of pooled vs independent levels, including inflation magnitude and finer cell-type stratification, see the Section 4.1 statistical supplement.
2. Dataset Catalog
2.1 Datasets and Scale [detailed analytics]
| # | Dataset | Type | Cells/Samples | Donors | Cell Types | Reference |
|---|---|---|---|---|---|---|
| 1 | GTEx | Bulk RNA-seq | 19,788 samples | 946 donors | — | GTEx Consortium, v11 |
| 2 | TCGA | Bulk RNA-seq | 11,069 samples | 10,274 donors | — | TCGA PanCancer |
| 3 | CIMA | scRNA-seq | 6,484,974 | 421 donors | 27 L2 / 100+ L3 | J. Yin et al., Science, 2026 |
| 4 | Inflammation Atlas Main | scRNA-seq | 4,918,140 | 817 samples* | 66+ | Jimenez-Gracia et al., Nature Medicine, 2026 |
| 5 | Inflammation Atlas Val | scRNA-seq | 849,922 | 144 samples* | 66+ | Validation cohort |
| 6 | Inflammation Atlas Ext | scRNA-seq | 572,872 | 86 samples* | 66+ | External cohort |
| 7 | scAtlas Normal | scRNA-seq | 2,293,951 | 317 donors | 102 subCluster | Q. Shi et al., Nature, 2025 |
| 8 | scAtlas Cancer | scRNA-seq | 4,146,975 | 717 donors (601 tumor-only) | 162 cellType1 | Q. Shi et al., Nature, 2025 |
| 9 | parse_10M | scRNA-seq | 9,697,974 | 12 donors × 90 cytokines (+PBS control) | 18 PBMC types | Oesinghaus et al., bioRxiv, 2026 |
Grand total: ~29 million single cells + ~31K bulk samples from 6 independent studies (9 datasets), 100+ cell types.
* Inflammation Atlas does not provide donor-level identifiers; the 817/144/86 values are sample counts. The donor–sample relationship is unknown, so correlations use sampleID as the independent unit.
2.2 Disease and Condition Categories
CIMA (421 healthy donors): Healthy population atlas with paired blood biochemistry (19 markers: ALT, AST, glucose, lipid panel, etc.) and plasma metabolomics (1,549 features). Enables age, BMI, sex, and smoking correlations with cytokine activity.
Inflammation Atlas (20 diseases): RA, SLE, Sjogren's, PSA, Crohn's, UC, COVID-19, Sepsis, HIV, HBV, BRCA, CRC, HNSCC, NPC, COPD, Cirrhosis, MS, Asthma, Atopic Dermatitis
scAtlas Normal (317 donors): 35 organs, 12 tissues with ≥20 donors for per-organ stratification (Breast 124, Lung 97, Colon 65, Heart 52, Liver 43, etc.)
scAtlas Cancer (717 donors, 601 tumor-only): 29 cancer types, 11 with ≥20 tumor-only donors for per-cancer stratification (HCC 88, PAAD 58, CRC 51, ESCA 48, HNSC 39, LUAD 36, NPC 36, KIRC 31, BRCA 30, ICC 29, STAD 27)
parse_10M: 90 cytokines × 12 donors — independent in vitro perturbation dataset for comparison. A considerable portion of cytokines (~58%) are produced in E. coli, with the remainder from insect (Sf21, 12%) and mammalian (CHO, NS0, HEK293, ~30%) expression systems. Because exogenous perturbagens may induce effects differing from endogenously produced cytokines, parse_10M serves as an independent comparison rather than strict ground truth. CytoSig/SecAct has a potential advantage in this regard, as it infers relationships directly from physiologically relevant samples.
2.3 Signature Matrices
| Matrix | Targets | Construction | Reference |
|---|---|---|---|
| CytoSig | 43 cytokines | Median log2FC across all experimental bulk RNA-seq | Jiang et al., Nature Methods, 2021 |
| LinCytoSig | 178 (45 cell types × 1–13 cytokines) | Cell-type-stratified median from CytoSig database (methodology) | This work |
| SecAct | 1,170 secreted proteins | Median global Moran's I across 1,000 Visium datasets | Ru et al., Nature Methods, 2026 (in press) |
3. Scientific Value Proposition
3.1 What Makes CytoAtlas Different from Deep Learning Approaches?
Most single-cell analysis tools use complex models (VAEs, GNNs, transformers) that produce aggregated, non-linear representations difficult to interpret biologically. CytoAtlas takes the opposite approach:
| Property | CytoAtlas (Ridge Regression) | Typical DL Approach |
|---|---|---|
| Model | Linear (z = Xβ + ε) | Non-linear (multi-layer NN) |
| Interpretability | Every gene's contribution is a coefficient | Feature importance approximated post-hoc |
| Conditionality | Activity conditional on specific gene set | Latent space mixes all features |
| Confidence | Permutation-based z-scores with CI | Often point estimates only |
| Generalization | Tested across 6 independent cohorts | Often held-out splits of same cohort |
| Bias | Transparent — limited by signature matrix genes | Hidden in architecture and training data |
The key insight: CytoAtlas is not trying to replace DL-based tools. It provides an orthogonal, complementary signal that a human scientist can directly inspect. When CytoAtlas says "IFNG activity is elevated in CD8+ T cells from RA patients," you can verify this by checking the IFNG signature genes in those cells.
3.2 What Scientific Questions Does CytoAtlas Answer?
- Which cytokines are active in which cell types across diseases? — IL1B/TNFA in monocytes/macrophages, IFNG in CD8+ T and NK cells, IL17A in Th17, VEGFA in endothelial/tumor cells, TGFB family in stromal cells — quantified across 20 diseases, 35 organs, and 15 cancer types.
- Are cytokine activities consistent across independent cohorts? — Yes. IL1B, TNFA, VEGFA, and TGFB family show consistent positive correlations across all 6 validation datasets (Figure 8).
- Does cell-type-specific biology matter for cytokine inference? — For select immune types, yes: LinCytoSig improves prediction for Basophils (+0.21 Δρ), NK cells (+0.19), and DCs (+0.18), but global CytoSig wins overall (Figures 11–12).
- Which secreted proteins beyond cytokines show validated activity? — SecAct (1,170 targets) achieves the highest correlations across all datasets (median ρ=0.33–0.49), with novel validated targets like Activin A (ρ=0.98), CXCL12 (ρ=0.92), and BMP family (Figure 13).
- Can we predict treatment response from cytokine activity? — We are incorporating cytokine-blocking therapy outcomes from bulk RNA-seq to test whether predicted cytokine activity associates with therapy response. Additionally, Inflammation Atlas responder/non-responder labels enable treatment response prediction using cytokine activity profiles as features.
3.3 Validation Philosophy
CytoAtlas validates against a simple but powerful principle: if CytoSig predicts high IFNG activity for a sample, that sample should have high IFNG gene expression. This expression-activity correlation is computed via Spearman rank correlation across donors/samples.
This is a conservative validation — it only captures signatures where the target gene itself is expressed. Signatures that act through downstream effectors would not be captured, meaning our validation underestimates true accuracy.
4. Validation Results
4.1 Overall Performance Summary [Full Details]
PRIMARY independent level: The summary table above reports results at each dataset’s PRIMARY independent level — the aggregation level where samples are fully independent (each donor counted once). This ensures correlation statistics are not inflated by donor duplication. See the “Primary Level” column for each dataset’s level.
How “N Targets” is determined: A target is included in the validation for a given atlas only if (1) the target’s signature genes overlap sufficiently with the atlas gene expression matrix, and (2) the target gene itself is expressed in enough samples to compute a meaningful Spearman correlation. Targets whose gene is absent or not detected in a dataset are excluded. CytoSig defines 43 cytokines and SecAct defines 1,170 secreted proteins. Inflammation Atlas Main retains only 33 of 43 CytoSig targets and 805 of 1,170 SecAct targets because 10 cytokine genes (BDNF, BMP4, CXCL12, GCSF, IFN1, IL13, IL17A, IL36, IL4, WNT3A) are not sufficiently expressed in these blood/PBMC samples.
Stratified levels (GTEx by_tissue, TCGA primary_by_cancer): Correlations are computed within each tissue/cancer type (ensuring independence), then summarized across groups. N Targets counts unique targets at the “all” aggregate level. Finer per-tissue or per-cancer breakdowns are available in Section 4.3 below.
4.2 Cross-Dataset Comparison: CytoSig vs SecAct [Statistical Methods]
Why does SecAct appear to underperform CytoSig in Inflammation Atlas Main?
This is a composition effect, not a genuine performance gap, confirmed by two complementary statistical tests:
Total comparison (Mann–Whitney U test): Compares the full ρ distributions of CytoSig (43 cytokine signatures) vs SecAct (~1,170 secreted protein signatures) using independence-corrected values. For GTEx/TCGA, each target’s representative ρ is the median across per-tissue/cancer values (median-of-medians); for other datasets, donor_only/tumor_only ρ is used directly. SecAct achieves a significantly higher median ρ in 5 of 6 datasets (GTEx: p = 4.76 × 10−4; TCGA: p = 2.85 × 10−3; CIMA: p = 3.18 × 10−2; scAtlas Normal: p = 1.04 × 10−4; scAtlas Cancer: p = 1.06 × 10−5). Inflammation Atlas Main is the sole exception (U = 14,101, p = 0.548, not significant) and the only dataset where CytoSig’s median ρ (0.323) exceeds SecAct’s (0.173).
Matched comparison (Wilcoxon signed-rank test): Restricts to the 32 targets shared between both methods (22 direct + 10 alias-resolved), each target serving as its own control. SecAct’s median ρ is consistently higher across all 6 datasets, reaching significance in 5 (GTEx: p = 3.54 × 10−5; TCGA: p = 3.24 × 10−6; CIMA: p = 2.28 × 10−2; scAtlas Normal: p = 3.54 × 10−5; scAtlas Cancer: p = 3.54 × 10−5). Inflammation Atlas Main is not significant (p = 0.141).
Inflammation Atlas Main is largely blood-derived, so many SecAct targets that perform well in multi-organ contexts contribute near-zero or negative correlations here. In fact, 99 SecAct targets are negative only in Inflammation Atlas Main but positive in all other datasets, reflecting tissue-specific expression limitations rather than inference failure. The “Matched” tab above demonstrates the fair comparison on equal footing.
4.3 Per-Tissue and Per-Cancer Stratified Validation [Statistical Methods]
Stratified validation: Instead of aggregating tissues/cancers into a single median-of-medians, this view shows the CytoSig vs SecAct comparison within each individual tissue (GTEx) or cancer type (TCGA). Mann-Whitney U test (Total tab: all targets) and Wilcoxon signed-rank test (Matched tab: 32 shared targets) with BH-FDR correction across all strata within each dataset.
Key insight: On matched targets (23 pairs per GTEx tissue, 19–21 per TCGA cancer after expression filtering), SecAct wins direction in 29/29 GTEx tissues and 32/33 TCGA cancer types, with 23/29 and 31/33 reaching significance (q<0.05). The sole exception is Acute Myeloid Leukemia (Δ=−0.08, q=0.83), a low-sample liquid tumor. This near-unanimous result across 61 of 62 strata rules out Simpson’s paradox. On total targets, SecAct wins in 28/29 GTEx tissues (21 significant) and 30/33 TCGA cancers (15 significant); the few CytoSig-favored strata (Brain in GTEx; Kidney Chromophobe, Ovarian, Uveal Melanoma in TCGA) are all non-significant. Since SecAct outperforms CytoSig on the same matched cytokines, the advantage is not about target breadth but about signature quality. SecAct’s spatial correlation signatures capture tissue-context-dependent regulation that CytoSig’s median log2FC signatures miss. The advantage is largest in tissues with complex cellular microenvironments (GTEx: Small Intestine Δ=+0.55, Vagina +0.34; TCGA: Pancreatic +0.34, Head & Neck +0.27) and smallest in homogeneous or low-cytokine contexts (GTEx: Heart +0.06, Skin +0.06; TCGA: Uveal Melanoma +0.07, Kidney Chromophobe +0.10).
4.4 Cross-Platform Comparison: Bulk vs Pseudobulk [Statistical Methods]
Cross-platform concordance: This section tests whether expression–activity relationships replicate across measurement technologies. For each tissue (GTEx) or cancer type (TCGA), we compute per-target Spearman ρ from bulk RNA-seq data and compare it to the same target’s ρ from single-cell pseudobulk data (scAtlas). Wilcoxon signed-rank tests (paired by target) with BH-FDR correction assess whether ρ values differ between platforms.
Key finding: Using all targets, SecAct shows significant bulk–pseudobulk differences in most strata (11/13 GTEx tissues, 5/11 TCGA cancers), while CytoSig shows almost none (1/13, 0/11). However, the Matched tabs reveal this is a statistical power effect, not a signal quality difference: when restricted to the same 32 shared targets, both CytoSig and SecAct show no significant platform differences (0/13 and 0/13 for GTEx; 0/11 and 1/11 for TCGA). The apparent platform sensitivity in SecAct’s full panel is a statistical power effect, not a signal quality difference: matched and unmatched SecAct targets show the same per-target platform shift (mean |Δ| = 0.298 vs 0.302, Mann–Whitney p = 0.82), but SecAct’s ~1,000 paired targets per tissue provide 25× more observations than CytoSig’s ~40, easily detecting the same tiny systematic shift (Δ ≈ 0.03) that CytoSig lacks power to detect. Core cytokine targets are platform-robust.
4.5 Best and Worst Correlated Targets
All 32 matched targets have data in ≥4 datasets. Using donor-level aggregation (matching Figure 5), two categories emerge (mean ρ across datasets):
| Consistent in Both CytoSig and SecAct (16 targets, both mean ρ > 0.25) | ||||
|---|---|---|---|---|
| Target | CytoSig Mean ρ | Range | SecAct Mean ρ | Datasets ρ>0.2 |
| IL1B | +0.56 | +0.24 to +0.72 | +0.58 | 6/6 & 6/6 |
| TNFA | +0.50 | +0.26 to +0.63 | +0.46 | 6/6 & 6/6 |
| IFNG | +0.44 | +0.30 to +0.62 | +0.35 | 6/6 & 5/6 |
| IL1A | +0.43 | +0.03 to +0.71 | +0.33 | 5/6 & 5/6 |
| IL27 | +0.43 | +0.19 to +0.56 | +0.40 | 5/6 & 5/6 |
| TGFB3 | +0.39 | +0.19 to +0.53 | +0.42 | 5/6 & 5/6 |
| IL6 | +0.38 | +0.17 to +0.53 | +0.48 | 5/6 & 6/6 |
| OSM | +0.38 | +0.06 to +0.49 | +0.57 | 5/6 & 6/6 |
| LIF | +0.37 | +0.20 to +0.62 | +0.50 | 5/6 & 6/6 |
| IL10 | +0.35 | +0.06 to +0.55 | +0.56 | 5/6 & 5/6 |
| CXCL12 | +0.34 | +0.10 to +0.57 | +0.59 | 4/6 & 5/6 |
| TGFB1 | +0.34 | +0.05 to +0.56 | +0.41 | 4/6 & 5/6 |
| BMP4 | +0.33 | −0.02 to +0.61 | +0.43 | 4/6 & 5/6 |
| BMP2 | +0.31 | +0.19 to +0.41 | +0.45 | 5/6 & 6/6 |
| Activin A | +0.29 | +0.12 to +0.54 | +0.56 | 4/6 & 6/6 |
| GMCSF | +0.26 | +0.01 to +0.46 | +0.37 | 4/6 & 4/6 |
| SecAct-Only: CytoSig Near-Zero, SecAct Rescues (11 targets) | ||||
|---|---|---|---|---|
| Target | CytoSig Mean ρ | Range | SecAct Mean ρ | Δ |
| LTA | −0.02 | −0.33 to +0.26 | +0.53 | +0.55 |
| HGF | +0.06 | −0.29 to +0.40 | +0.58 | +0.51 |
| TWEAK | −0.02 | −0.22 to +0.11 | +0.44 | +0.47 |
| IL15 | +0.11 | −0.05 to +0.43 | +0.57 | +0.47 |
| BMP6 | +0.04 | −0.40 to +0.26 | +0.49 | +0.45 |
| TRAIL | −0.00 | −0.54 to +0.58 | +0.44 | +0.44 |
| CD40L | +0.02 | −0.55 to +0.57 | +0.46 | +0.43 |
| FGF2 | +0.04 | −0.23 to +0.29 | +0.46 | +0.42 |
| MCSF | +0.16 | −0.26 to +0.50 | +0.40 | +0.24 |
| IL21 | −0.02 | −0.22 to +0.09 | +0.22 | +0.24 |
| BDNF | +0.11 | −0.07 to +0.20 | +0.33 | +0.21 |
Remaining 5 targets do not fit either category: VEGFA and IFNL are CytoSig-only (CytoSig mean +0.38/+0.21, SecAct < 0.2), GDF11 and GCSF are borderline (CytoSig mean +0.23 each, SecAct +0.43/+0.35), and IL36 shows near-zero correlations in both methods.
Key insight: CytoSig’s median log2FC signatures reliably detect 16 canonical cytokines with strong transcriptional programs. SecAct’s spatial correlation signatures additionally rescue 11 targets where CytoSig averages near zero—including membrane-bound (CD40L, TRAIL), paracrine (HGF, FGF2), and heteromeric (LTA) signaling targets. TWEAK and IL21 are the clearest cases: CytoSig averages near zero across all datasets, while SecAct achieves +0.44 and +0.22. Note: this section uses donor-level aggregation (matching Figure 5), not the independence-corrected median-of-medians used in §4.1–4.3. Select “SecAct” in the dropdown above to explore interactively.
4.6 Cross-Atlas Consistency
| Tier | Targets | Pattern |
|---|---|---|
| Universal (ρ>0.1 in all 6) | IL1B, TNFA, IFNG, IL6, BMP2, VEGFA | Robust across all cohorts and platforms (mean ρ 0.31–0.58) |
| Mostly consistent (4–5 of 6) | IL10, TGFB1, CXCL12, GMCSF, HGF | Occasional near-zero outliers; TGFB1 slightly negative only in scAtlas Normal (−0.05) |
| Context-dependent (≤3 of 6) | IL4, IL17A, EGF | Sign changes across cohorts; Th2/Th17 cytokines absent from Inflammation Main |
Key insight: No systematic bulk vs single-cell advantage exists—mean |ρ| across 14 targets: GTEx 0.26, TCGA 0.33, CIMA 0.26, Inflammation Main 0.44, scAtlas Normal 0.36, scAtlas Cancer 0.37. Variation is target-specific and cohort-specific rather than platform-driven. The universal tier (IL1B, TNFA, IFNG, IL6) represents cytokine axes fundamental to inflammation biology, while context-dependent targets (IL4, IL17A) reflect Th2/Th17 pathways active only in specific disease cohorts.
4.7 Effect of Aggregation Level [Statistical Methods]
Aggregation levels explained: Pseudobulk profiles are aggregated at increasingly fine cell-type resolution. At coarser levels, each pseudobulk profile averages more cells, yielding smoother expression estimates but masking cell-type-specific signals. At finer levels, each profile is more cell-type-specific but based on fewer cells.
Key insight: Beyond ~L2 annotation depth, correlations become noise-dominated—cell-type stratification fragments the donor pool below the sample size needed for reliable correlation (signal/null ratio <2×, % positive → 50%). The practical limit is samples-per-group, not annotation depth: Inflammation Main retains signal at L2 (553 samples/group) while scAtlas Normal is marginal at its shallowest stratification (22 samples/group). In aggregate, SecAct’s larger target pool (~1,170) shows higher retention than CytoSig (43 targets), but on matched targets the advantage disappears.
| Atlas | Level | Description | N Cell Types |
|---|---|---|---|
| CIMA | Donor Only | Whole-sample pseudobulk per donor | 1 (all) |
| Donor × L1 | Broad lineages (B, CD4_T, CD8_T, Myeloid, NK, etc.) | 7 | |
| Donor × L2 | Intermediate (CD4_memory, CD8_naive, DC, Mono, etc.) | 28 | |
| Donor × L3 | Fine-grained (CD4_Tcm, cMono, Switched_Bm, etc.) | 39 | |
| Donor × L4 | Finest marker-annotated (CD4_Th17-like_RORC, cMono_IL1B, etc.) | 73 | |
| Inflammation Atlas Main | Donor Only | Whole-sample pseudobulk per donor | 1 (all) |
| Donor × L1 | Broad categories (B, DC, Mono, T_CD4/CD8 subsets, etc.) | 18 | |
| Donor × L2 | Fine-grained (Th1, Th2, Tregs, NK_adaptive, etc.) | 65 | |
| scAtlas Normal | Donor × Organ | Per-organ pseudobulk (Bladder, Blood, Breast, Lung, etc.) | 25 organs |
| Donor × Organ × CT1 | Broad cell types within each organ | 191 | |
| Donor × Organ × CT2 | Fine cell types within each organ | 356 | |
| scAtlas Cancer | All Tumor Cells | Whole-sample pseudobulk per tumor donor | 1 (all) |
| Per Cancer Type | Pseudobulk stratified by cancer type (HCC, PAAD, CRC, etc.) | 29 types | |
| Per Cancer Type × CT1 | Broad cell types within each cancer type | ~120 |
4.8 Representative Scatter Plots
4.9 Biologically Important Targets Heatmap
How each correlation value is computed: For each (target, atlas) cell, we compute Spearman rank correlation between predicted cytokine activity (ridge regression z-score) and target gene expression across all donor-level pseudobulk samples. Specifically:
- Pseudobulk aggregation: For each atlas, gene expression is aggregated to the donor level (one profile per donor or donor × cell type).
- Activity inference: Ridge regression (
secactpy.ridge, λ=5×105) is applied using the signature matrix (CytoSig: 4,881 genes × 43 cytokines; SecAct: 7,919 genes × 1,170 targets) to predict activity z-scores for each pseudobulk sample. - Correlation: Spearman ρ is computed between the predicted activity z-score and the original expression of the target gene across all donor-level samples within that atlas. A positive ρ means higher predicted activity tracks with higher target gene expression.
GTEx uses per-tissue pseudobulk (median-of-medians across 29 tissues); TCGA uses per-cancer type (median-of-medians across 33 cancers); CIMA/Inflammation Atlas Main use donor-only; scAtlas Normal uses donor-only; scAtlas Cancer uses tumor-only.
4.10 Per-Target Correlation Rankings
5. CytoSig vs LinCytoSig vs SecAct Comparison
5.1 Method Overview
| Method | Targets | Genes | Specificity | Selection |
|---|---|---|---|---|
| CytoSig | 43 cytokines | 4,881 curated | Global (all cell types) | — |
| LinCytoSig (orig) | 178 (45 CT × cytokines) | All ~20K | Cell-type specific | Matched cell type |
| LinCytoSig (gene-filtered) | 178 | 4,881 (CytoSig overlap) | Cell-type specific | Matched cell type |
| LinCytoSig Best (combined) | 43 (1 per cytokine) | All ~20K | Best CT per cytokine | Max combined GTEx+TCGA ρ |
| LinCytoSig Best (comb+filt) | 43 (1 per cytokine) | 4,881 (CytoSig overlap) | Best CT per cytokine | Max combined ρ (filtered) |
| LinCytoSig Best (GTEx) | 43 (1 per cytokine) | All ~20K | Best CT per cytokine | Max GTEx ρ |
| LinCytoSig Best (TCGA) | 43 (1 per cytokine) | All ~20K | Best CT per cytokine | Max TCGA ρ |
| LinCytoSig Best (GTEx+filt) | 43 (1 per cytokine) | 4,881 (CytoSig overlap) | Best CT per cytokine | Max GTEx ρ (filtered) |
| LinCytoSig Best (TCGA+filt) | 43 (1 per cytokine) | 4,881 (CytoSig overlap) | Best CT per cytokine | Max TCGA ρ (filtered) |
| SecAct | 1,170 secreted proteins | Spatial Moran’s I | Global (all cell types) | — |
Gene filter: LinCytoSig signatures restricted from ~20K to CytoSig’s 4,881 curated genes. Best selection: For each cytokine, test all cell-type-specific LinCytoSig signatures and select the one with the highest bulk RNA-seq correlation. “Combined” uses pooled GTEx+TCGA; “GTEx” and “TCGA” select independently per bulk dataset. “+filt” variants apply the same cell-type selection but restrict to CytoSig gene space. See LinCytoSig Methodology for details.
Ten methods compared on identical matched pairs across 4 combined datasets:
- CytoSig — 43 cytokines, 4,881 curated genes, global (all cell types)
- LinCytoSig (orig) — cell-type-matched signatures, all ~20K genes
- LinCytoSig (gene-filtered) — cell-type-matched signatures, restricted to CytoSig’s 4,881 genes
- LinCytoSig Best (combined) — best cell-type signature per cytokine (selected by combined GTEx+TCGA bulk ρ), all ~20K genes
- LinCytoSig Best (comb+filt) — best combined bulk signature, restricted to 4,881 genes
- LinCytoSig Best (GTEx) — best per cytokine selected by GTEx-only bulk ρ, all ~20K genes
- LinCytoSig Best (TCGA) — best per cytokine selected by TCGA-only bulk ρ, all ~20K genes
- LinCytoSig Best (GTEx+filt) — GTEx-selected best, restricted to 4,881 genes
- LinCytoSig Best (TCGA+filt) — TCGA-selected best, restricted to 4,881 genes
- SecAct — 1,170 secreted proteins (Moran’s I), subset matching CytoSig targets
Key findings:
- SecAct achieves the highest median ρ across all 4 combined datasets, benefiting from spatial-transcriptomics-derived signatures.
- CytoSig outperforms most LinCytoSig variants at donor level, with one notable exception: scAtlas Normal Best-orig (0.298) exceeds CytoSig (0.216).
- Gene filtering improves LinCytoSig in most datasets (CIMA +102%, Inflammation Atlas Main), confirming noise reduction from restricting the gene space.
- GTEx-selected best performs comparably to combined-selected in most datasets but slightly better in scAtlas Cancer (0.300 vs 0.275). TCGA-selected best generally underperforms other selection strategies, suggesting GTEx’s broader tissue coverage provides more generalizable selections.
- Gene filtering of GTEx/TCGA-selected: GTEx+filt and TCGA+filt show mixed results — filtering sometimes improves (e.g., TCGA+filt in Inflammation Atlas Main: 0.260 vs TCGA-orig 0.168) but can also reduce performance, indicating the optimal gene space depends on both the selection dataset and target dataset context.
- General ranking: SecAct > CytoSig > LinCytoSig Best variants > LinCytoSig (filt) > LinCytoSig (orig), though dataset-specific exceptions exist.
For detailed issue analysis including selection circularity, case studies (IFNG, TGFB1, IL1B), and cross-validation design, see the LinCytoSig Open Issues & Analysis companion document.
5.2 Effect of Aggregation Level
Methodology: At each cell-type aggregation level (CIMA: L1–L4 = 7–73 cell types; Inflammation: L1–L2; scAtlas: CT1–CT2 = coarse/fine), we match CytoSig, LinCytoSig, and SecAct on identical (cytokine, cell type) pairs — using the exact same pseudobulk samples and identical n for all three methods. For each pair, Spearman ρ measures agreement between predicted activity and target gene expression. If lineage-specific aggregation helps, LinCytoSig should increasingly outperform CytoSig as cell-type resolution increases (L1 → L4).
5.2.1 Distribution at Each Level
5.2.2 Summary
n = number of three-way matched pairs. Δρ = LinCytoSig − competitor (negative = LinCytoSig underperforms).
5.2.3 Which Cell Types Benefit?
Aggregated across all datasets at finest celltype level. Green = LinCytoSig wins more; red = LinCytoSig loses more.
5.2.4 Which Cytokines Benefit?
Sorted by mean Δρ vs CytoSig (best to worst).
Key finding: Lineage-specific aggregation provides no systematic advantage at any level.
- At every level, LinCytoSig underperforms CytoSig (mean Δρ ranges from −0.08 at coarse L1 to −0.02 at fine L4 in CIMA). Finer cell types reduce the gap slightly but never close it.
- SecAct wins at every level in CIMA and scAtlas. In Inflammation Atlas Main L2, LinCytoSig is nearly tied with SecAct (Δρ = +0.01) but still loses to CytoSig.
- Per cell type: Only 5 of 43 cell types show consistent LinCytoSig advantage vs CytoSig (NK Cell, Basophil, DC, Trophoblast, Arterial Endothelial). No cell type beats SecAct.
- Interpretation: CytoSig’s global signature, derived from median log2FC across all cell types, already captures the dominant transcriptional response. Restricting to a single cell type’s response introduces noise from small sample sizes without gaining meaningful lineage specificity. The hypothesis that finer resolution should favor LinCytoSig is not supported by the data.
See Issue 3: Celltype-Level Evaluation for a matched celltype-level analysis where LinCytoSig is tested on its intended target (e.g., Fibroblast__TGFB1 on fibroblast pseudobulk).
5.3 SecAct: Breadth Over Depth
- Highest median ρ in single-cell datasets (scAtlas Normal: 0.455, Cancer: 0.399, independence-corrected)
- Highest median ρ in bulk RNA-seq (GTEx: 0.314, TCGA: 0.357, independence-corrected median-of-medians)
- 95.8% positive correlation in TCGA (independence-corrected)
- Wins decisively at celltype level against both CytoSig and LinCytoSig in scAtlas (19/3 wins vs CytoSig in scAtlas Normal, 20/2 in Cancer)
6. Key Takeaways for Scientific Discovery
6.1 What CytoAtlas Enables
- Quantitative cytokine activity per cell type per disease — 43 CytoSig cytokines + 1,170 SecAct secreted proteins across 29M cells
- Cross-disease comparison — same signatures validated across 20 diseases, 35 organs, 15 cancer types
- Independent perturbation comparison — parse_10M provides 90 cytokine perturbations × 12 donors × 18 cell types for independent comparison with CytoSig predictions
- Multi-level validation — donor, donor × celltype, bulk RNA-seq (GTEx/TCGA), and resampled bootstrap validation across 6 datasets
6.2 Limitations
- Linear model: Cannot capture non-linear cytokine interactions
- Transcriptomics-only: Post-translational regulation invisible
- Signature matrix bias: Underrepresented cell types have weaker signatures
- Validation metric: Expression-activity correlation underestimates true accuracy (signatures acting through downstream effectors are not captured)
6.3 Future Directions
- scGPT cohort integration (~35M cells)
- cellxgene Census integration
- Classification of cytokine blocking therapy
7. Appendix: Technical Specifications
A. Computational Infrastructure
- GPU: NVIDIA A100 80GB (SLURM gpu partition)
- Memory: 256–512GB host RAM per node
- Pipeline: 24 Python scripts, 18 pipeline subpackages (~18.7K lines)
- API: 262 REST endpoints across 17 routers
- Frontend: 12 pages, 122 source files, 11.4K LOC
B. Statistical Methods
- Activity inference: Ridge regression (λ=5×105, z-score normalization, permutation-based significance)
- Correlation: Spearman rank correlation
- Multiple testing: Benjamini-Hochberg FDR (q < 0.05)
- Bootstrap: 100–1000 resampling iterations
- Differential: Wilcoxon rank-sum test with effect size
C. Detailed System Architecture
A comprehensive technical reference covering both cytoatlas-pipeline (offline GPU computation) and cytoatlas-api (online serving) is available as a separate document.